Abstract: The apparent source of sound from a plurality of transducers arrayed around a listening area is automatically controlled by a prestored pattern sequencer and a series of variable voltage levels. The prestored patterns reflect switch connections to the transducers which are energized by varying signals so that the sound appears to the listener to move relative to the transducers. In one embodiment, varying analog signal levels are switched under pattern control while, in another embodiment, the controls are entirely effected with digital circuitry up to the final analog output stages. The sequences of switch patterns can be automatically selected or operator selected and can provide the effect of rotating sound or any of a wide variety of prestored patterns.